CN103999073A - Engine control device - Google Patents
Engine control device Download PDFInfo
- Publication number
- CN103999073A CN103999073A CN201180075464.3A CN201180075464A CN103999073A CN 103999073 A CN103999073 A CN 103999073A CN 201180075464 A CN201180075464 A CN 201180075464A CN 103999073 A CN103999073 A CN 103999073A
- Authority
- CN
- China
- Prior art keywords
- core
- lattice
- engine
- operating condition
- lattice point
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
- 230000009471 action Effects 0.000 claims abstract description 9
- 239000011159 matrix material Substances 0.000 claims abstract description 7
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 claims description 18
- 230000006870 function Effects 0.000 claims description 5
- 238000000034 method Methods 0.000 description 29
- 238000012545 processing Methods 0.000 description 18
- 238000004891 communication Methods 0.000 description 12
- 230000008859 change Effects 0.000 description 10
- 230000008569 process Effects 0.000 description 10
- 238000010586 diagram Methods 0.000 description 6
- 238000005516 engineering process Methods 0.000 description 4
- 239000012141 concentrate Substances 0.000 description 3
- 238000013459 approach Methods 0.000 description 2
- 230000008901 benefit Effects 0.000 description 2
- 230000005540 biological transmission Effects 0.000 description 1
- 238000012790 confirmation Methods 0.000 description 1
- 239000002283 diesel fuel Substances 0.000 description 1
- 230000010365 information processing Effects 0.000 description 1
- 230000008450 motivation Effects 0.000 description 1
- 230000006911 nucleation Effects 0.000 description 1
- 238000010899 nucleation Methods 0.000 description 1
- 238000005457 optimization Methods 0.000 description 1
- 239000004065 semiconductor Substances 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
- 238000013316 zoning Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
- G05B15/00—Systems controlled by a computer
- G05B15/02—Systems controlled by a computer electric
-
- F—MECHANICAL ENGINEERING; LIGHTING; HEATING; WEAPONS; BLASTING
- F02—COMBUSTION ENGINES; HOT-GAS OR COMBUSTION-PRODUCT ENGINE PLANTS
- F02D—CONTROLLING COMBUSTION ENGINES
- F02D41/00—Electrical control of supply of combustible mixture or its constituents
- F02D41/24—Electrical control of supply of combustible mixture or its constituents characterised by the use of digital means
- F02D41/2406—Electrical control of supply of combustible mixture or its constituents characterised by the use of digital means using essentially read only memories
- F02D41/2409—Addressing techniques specially adapted therefor
- F02D41/2416—Interpolation techniques
-
- F—MECHANICAL ENGINEERING; LIGHTING; HEATING; WEAPONS; BLASTING
- F02—COMBUSTION ENGINES; HOT-GAS OR COMBUSTION-PRODUCT ENGINE PLANTS
- F02D—CONTROLLING COMBUSTION ENGINES
- F02D41/00—Electrical control of supply of combustible mixture or its constituents
- F02D41/24—Electrical control of supply of combustible mixture or its constituents characterised by the use of digital means
- F02D41/26—Electrical control of supply of combustible mixture or its constituents characterised by the use of digital means using computer, e.g. microprocessor
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F12/00—Accessing, addressing or allocating within memory systems or architectures
- G06F12/02—Addressing or allocation; Relocation
- G06F12/08—Addressing or allocation; Relocation in hierarchically structured memory systems, e.g. virtual memory systems
- G06F12/0802—Addressing of a memory level in which the access to the desired data or data block requires associative addressing means, e.g. caches
- G06F12/0806—Multiuser, multiprocessor or multiprocessing cache systems
- G06F12/084—Multiuser, multiprocessor or multiprocessing cache systems with a shared cache
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F9/00—Arrangements for program control, e.g. control units
- G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
- G06F9/46—Multiprogramming arrangements
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02T—CLIMATE CHANGE MITIGATION TECHNOLOGIES RELATED TO TRANSPORTATION
- Y02T10/00—Road transport of goods or passengers
- Y02T10/10—Internal combustion engine [ICE] based vehicles
- Y02T10/40—Engine management systems
Landscapes
- Engineering & Computer Science (AREA)
- General Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Mechanical Engineering (AREA)
- Software Systems (AREA)
- Chemical & Material Sciences (AREA)
- Combustion & Propulsion (AREA)
- Computer Hardware Design (AREA)
- Microelectronics & Electronic Packaging (AREA)
- Automation & Control Theory (AREA)
- Combined Controls Of Internal Combustion Engines (AREA)
- Multi Processors (AREA)
Abstract
多核处理器上以点阵状被配置的多个核中的至少一部分以1比1的方式，被与排列在以第一运转条件、第二运转条件以及第三运转条件作为轴的三维直角坐标系上的多个阵点建立关联，而且，所述多个核中的至少一部分上被分配有，用于对被建立了关联的阵点处的控制值进行运算的运算程序。在与阵点被建立了关联的各个核在当前的动作点所属的所述三维直角坐标系上的运转空间为，通过包含被与自身建立了关联的阵点在内的相邻的八个阵点而被定义的空间的情况下，通过所述的运算程序而对被与自身建立了关联的阵点处的控制值进行运算，并向插补计算核进行输送，而在上述以外的情况下，使控制值的运算休止。插补计算核利用包围当前的动作点的相邻的八个阵点处的控制值，而对所述动作点处的控制值进行插补计算。多核处理器将通过插补计算核而得到的当前的动作点处的控制值作为作动器的控制目标值而输出。
At least a part of the plurality of cores arranged in a dot matrix on the multi-core processor is arranged in a 1:1 manner on a three-dimensional Cartesian coordinate with the first operating condition, the second operating condition, and the third operating condition as axes A plurality of lattice points on the system are associated, and at least a part of the plurality of cores is allocated with an operation program for operating control values at the associated lattice points. The operating space of each core associated with a lattice point on the three-dimensional Cartesian coordinate system to which the current action point belongs is, through the eight adjacent lattice points including the lattice point associated with itself In the case of a space defined by points, the control value at the matrix point associated with itself is calculated by the above-mentioned calculation program, and sent to the interpolation calculation core, and in the case other than the above , to stop the operation of the control value. The interpolation calculation core performs interpolation calculation on the control value at the current action point by using the control values at the eight adjacent matrix points surrounding the current action point. The multi-core processor outputs the control value at the current operating point obtained by the interpolation calculation core as the control target value of the actuator.
Description
Technical field
The present invention relates to a kind of control device that decides the engine for automobile of the control desired value of actuator according to multiple operating conditions, specifically, relate to that a kind of utilization has the polycaryon processor of multiple cores and the engine control system that the control desired value of actuator carried out to computing.
Background technology
In recent years, on the semi-conductor chip of, be equipped with the arithmetic unit of multiple CPU cores, i.e. the use of polycaryon processor is suggested in various fields.For example disclosed in TOHKEMY 2008-269487 communique, the use of polycaryon processor in the field of the control of engine for automobile is just in studied.In TOHKEMY 2008-269487 communique, disclosed technology is the technology that is reduced to object of the power consumption to use polycaryon processor in engine control system in the situation that.According to this technology, in the stopped process of engine control, carry out specific control while processing, the pattern of microcomputer is set to the different low consumption power mode of pattern while action from engine.In low consumption power mode, the check figure that microcomputer uses is set as the check figure using while being less than engine action.In the stopped process of engine control, do not there is the generation power from alternator etc. as in engine course of action, can not implement the charging of on-vehicle battery.Thus, the check figure moving in the time of the specific execution of controlling processing is more, and the charging power of accumulator more can promptly be consumed.
One of advantage that uses polycaryon processor is that its processing power is higher.Compared with the single core processor that is with CPU core, polycaryon processor can obtain higher processing power.Processing power is higher becomes one of motivation polycaryon processor being used in engine control system.This be due to, be equipped in recent years the quantity of the actuator on engine and the cause that kind increases gradually.For the suitably running of control engine, the control desired value of described actuator need to be set as to the suitable value corresponding to the operating condition of engine.For this reason, although in existing engine control system, implement optimization computing, and computing load now can more increase when more in the number of actuator or kind.In the processing power of the single core processor using in existing engine control system, there is soon not enough possibility in existence for the computing load increasing gradually.Use polycaryon processor for thering is huge advantage as being contemplated to computing load from now on for the field of engine control increasing gradually.
But the processing power of polycaryon processor is not merely only decided by check figure.Need to make multiple core move efficiently in order to improve processing power, require to study on the software that makes core action for this reason.For example, polycaryon processor can to multiple core allocating tasks and to described tasks in parallel process, can make in overall processing power to produce larger difference according to the situation of the method for this parallelization, and also will produce larger difference on power consumption.
Formerly technical literature
Patent documentation
Patent documentation 1: TOHKEMY 2008-269487 communique
Patent documentation 2: TOHKEMY 2011-053876 communique
Patent documentation 3: TOHKEMY 2008-197776 communique
Non-patent literature
Non-patent literature 1: hole Jiang Jianzhi, Lin Jianyi, " in torus network, the best is entirely to full communication mode ", information processing association collection of thesis, vol.34, no.4, PP628-637,1993.
Summary of the invention
Invent problem to be solved
Problem of the present invention is, makes the control desired value of the related one or more actuator of engine control carry out computing with high speed and low consumption electric power by polycaryon processor.And, in order to reach this problem, the invention provides a kind of engine control system as described below.
The invention provides a kind of engine control system, it has the polycaryon processor that disposes multiple cores with lattice-like.Although alleged dot matrix refers to plane lattice herein, its figure is not defined.Except the general square lattice or rectangular lattice of the configuration figure as core, also can adopt other the figure such as rhombohedral lattice.
On the multiple cores that are equipped in polycaryon processor, with 1 in 1 mode and be associated with the multiple lattice points on being arranged on using the first operating condition, the second operating condition and the 3rd operating condition as the three-dimensional cartesian coordinate system of axle.It is associated that the first operating condition and the second operating condition are set up the operational ton of actuator and the controlled quentity controlled variable of engine, because of but the operating condition of particular importance, an one typical example is the combination of engine speed and engine loading.Although the 3rd operating condition is condition arbitrarily, also can be made as should be by preferential operating condition under the operating condition of current engine.For example, in the warming-up of engine, using engine water temperature as the 3rd operating condition, and after the end of warming-up, the 3rd operating condition also can change to the speed of a motor vehicle or intake air temperature.
Being established associated core with the lattice point on three-dimensional cartesian coordinate system can not be also the whole core being equipped in polycaryon processor.For example, in the case of core is configured in the rectangular lattice of N × M, also can only make to be established associated as the core of the zoning of its a part of n × m with lattice point.Be established with lattice point on each associated core and allocated in advance, for the operation program that the controlling value of each actuator that has been established associated lattice point place is carried out to computing.In the situation that each core has local storage, the controlling value operation program being assigned with can be stored in described local storage.In the case, on the local storage of each core, can also store the interpolation calculation procedure for carrying out interpolation calculating described later.Below, core among the core being equipped on polycaryon processor, that be assigned with controlling value operation program is called to controlling value operation core.In addition, controlling value operation program is preferably, and by Optimal Control, Optimal Control value is carried out the program of computing.
On described three-dimensional cartesian coordinate system, multiple apron spaces are defined by the multiple lattice points that are arranged thereon.Because the three-dimensional dot pattern being obtained by lattice point on three-dimensional cartesian coordinate system is simple cubic lattice, thereby define an apron space by the adjacent lattice point of eight.By the determined current operating point of each currency of each operating condition, belong to the some apron spaces on three-dimensional cartesian coordinate system.
Each controlling value operation core is programmed as follows,, apron space under current operating point is, set up with self the space that associated lattice point is defined at interior adjacent eight lattice points by comprising, carried out computing by described operation program to set up the controlling value at associated lattice point place with self, and calculate core to interpolation and carry.In addition, each controlling value operation core is programmed as follows,, set up the space that associated lattice point is defined at interior adjacent eight lattice points the computing of the controlling value of having been set up associated lattice point place with self of stopping with self by comprising at the apron space under current operating point.In addition, be whether the confirmation of having been set up the space that adjacent eight lattice points of associated lattice point are defined by comprising with self for the apron space under current operating point, in each controlling value operation core for example, with the regular time step control cycle of each engine (with) and be implemented.
Interpolation is calculated core and is programmed as follows, that is, utilize the controlling value at adjacent eight lattice point places that the apron space under current operating point is defined, and the controlling value at current operating point place is carried out to interpolation calculating.Polycaryon processor will be examined the interpolation of executing and calculates the controlling value at the current operating point place obtaining and export as the control desired value of actuator by being calculated by interpolation.
; according to the present invention; engine control system is by being established eight associated cores and the controlling value at each lattice point place is carried out to computing with adjacent eight lattice points that surround current operating point in three-dimensional mode, and according to described operation result, the controlling value at current operating point place carried out to interpolation calculating.According to this mode, due to core by separately of the computing of the controlling value at each lattice point place and implemented concurrently, thereby the needed time of computing of the control desired value of actuator is shortened significantly compared with utilizing the situation of existing single core processor.In addition,, because engine control system is for being stopped the computing of controlling value by the core beyond having set up eight associated cores with described adjacent eight lattice points, therefore can and suppress power consumption when the control desired value to actuator is carried out high-speed computation.
In addition, it is associated that the lattice point on so-called three-dimensional cartesian coordinate system and the core in plane both can have been set up randomly, also can set up according to fixing rule association.But, in the time the data of the controlling value calculating on each core being passed to interpolation calculating core, can produce the communication delay corresponding with its distance physically.If consider this point, be preferably according to following this rule and implement the associated of lattice point and core.
Preferably regular according to this, be set with the multiple core groups that formed by four adjacent cores.And it is associated that each core group and following lattice point are established, that is, and four different continuous lattice points of each coordinate coordinate identical and the 3rd operating condition of the first operating condition and the second operating condition.And, each core group using with the first operating condition and the second operating condition are arranged as the identical arrangement mode of the lattice point on the two-dimensional direct angle coordinate system of axle.As by rule to implement lattice point associated with the foundation of core, because the controlling value at each lattice point place is carried out computing by eight cores that physically adjoin each other, can reduce communication delay poor at internuclear place.Because the difference of communication delay is less, can carry out efficiently computing as processor entirety.
In addition, lattice point and core associated, more preferably, lattice point responsible each core is once substituted in core group in the case of implementing by above-mentioned this rule.Accordingly, can prevent because computing load concentrates on the deteriorated of the core that causes on a part of core.In addition, the alternative of the responsible lattice point of each core as implemented in each predetermined operation process.
The nuclear energy with interpolation calculating kernel function is enough fixed as any one core.In this case, both can, by the some interpolation calculating core that is set as among controlling value operation core, also can calculate core from selected interpolation among controlling value operation core core in addition.But, if implement the associated of lattice point and core according to above-mentioned this rule, be preferably and make to calculate core and bring into play core dynamically variation between controlling value operation core of function as interpolation.In this case, the apron space of each controlling value operation core under current operating point is, in situation by the space that is defined including adjacent eight lattice points of having been set up with self associated lattice point, be programmed to carry out the mode of following processing.
First, each controlling value operation core is to having been set up the representative lattice point among adjacent eight lattice points whether associated lattice point be equivalent to the apron space under current operating point to define and having been judged with self.Below, will be called and represent core corresponding to the core that represents lattice point.Each controlling value operation core self be not the in the situation that of representing core, using the controlling value of having been set up associated lattice point place with self to sending as other cores that represent core.On the other hand, self be the in the situation that of representing core, from operating point to current under other the lattice point that defines of apron space be established each associated core, receive the controlling value being gone out by these kernel operations.And the controlling value at adjacent eight the lattice point places that define according to the apron space under the operating point to current, carries out interpolation calculating by interpolation calculation procedure to the controlling value at current operating point place.Because each core is programmed in the mode of implementing this processing, thereby the poor of internuclear communication delay not only can be reduced, communication delay itself can also be reduced.
Brief description of the drawings
Fig. 1 is the figure that represents the arrangement of the core of the related polycaryon processor of embodiments of the present invention.
Fig. 2 is the figure that represents the summary of the hardware configuration of the related polycaryon processor of embodiments of the present invention.
Fig. 3 is the figure that represents the summary of the software configuration of the related polycaryon processor of embodiments of the present invention.
Fig. 4 is the figure of the arrangement of the lattice point on the three-dimensional cartesian coordinate system representing taking engine speed, engine loading and engine water temperature as axle.
Fig. 5 represents to be arranged on the associated figure between the multiple cores on multiple lattice points and the polycaryon processor on three-dimensional cartesian coordinate system.
Fig. 6 is for the multiple lattice points that are arranged on three-dimensional cartesian coordinate system are set up to the figure that associated method is elaborated to the multiple cores on polycaryon processor.
Fig. 7 is the figure describing for the operational method of the control desired value of the actuator to according to the embodiment of the present invention.
Fig. 8 is the figure describing for the operational method of the control desired value of the actuator to according to the embodiment of the present invention.
Fig. 9 is the process flow diagram that represents the Optimal Control value calculation procedure being performed by each core in embodiments of the present invention.
Figure 10 is the figure describing for the rotation method of the core distribution in the core group to implementing according to the embodiment of the present invention.
Figure 11 represents that the core being performed by each core in embodiments of the present invention distributes the process flow diagram of rotation program.
Figure 12 is the figure describing for the operational method of the control desired value of the actuator of the Change Example 1 to according to the embodiment of the present invention.
Figure 13 is the figure describing for the operational method of the control desired value of the actuator of the Change Example 2 to according to the embodiment of the present invention.
Embodiment
With reference to accompanying drawing, the engine control system of embodiments of the present invention is described.
Having applied engine control system of the present invention is, uses polycaryon processor and device that the control desired value of the related actuator of engine control is carried out to computing.To the not restriction of the kind of controlled engine for automobile and structure by engine control system, in addition, also do not limit for kind and the number of actuator in the present invention.For example, the engine control system of present embodiment can be petrolic control device.In the case, engine control system carries out computing to the control desired value of the actuator such as air throttle, portfire, thrower, EGR valve, variable valve actuator for air and exhaust gas by-pass valve.In addition, the engine control system of present embodiment can be also the control device of diesel motor.In the case, the control desired value of the actuator of engine control system to thrower, EGR valve, diesel oil air throttle and variable nozzle turbine etc. is carried out computing.
The figure of the arrangement of the core that Fig. 1 is the polycaryon processor that represents the engine control system of embodiments of the present invention and have.The related polycaryon processor 100 of present embodiment is characterized as constructional, has and disposes multiple core 102 with lattice-like.The number of core 102 is more than 100, and polycaryon processor 100 is called as the processor with multiple cores.Fig. 2 is the figure that represents the summary of the hardware configuration of polycaryon processor 100.Each core 102 possesses CPU104 and the local storage 106 of subsidiary cache memory, between core 102, is joined together by bus.In local storage 106, store the various data that use when the various programs that are performed by CPU104 and this program are carried out.In addition, although omitted diagram, in bus 108, be also connected with at internuclear shared shared storage.
Fig. 3 is the figure that represents the summary of the software configuration of polycaryon processor 100.In polycaryon processor 100, on each core 102, different OS110 moves.In addition, on each OS110, different application program 112 is moved.Following Optimal Control value operation program or interpolation calculation procedure are in this application program 112.The OS110 associated with each core 102 and application program 112 are on each core 102 and be installed in local storage 106.In the polycaryon processor 100 with this software configuration, each core 102 executive utility 112 individually.In the case of the data that the operation result that need to be obtained by other core 102 for executive utility 112 or other core 102 have, carry out the exchange of executing data by the internuclear communication (communicating by letter between OS) of carrying out via bus 108.
The engine control system of present embodiment make the operational ton of actuator and the controlled quentity controlled variable of engine set up associated after, and obtain currency separately for three important operating conditions, and decide the control desired value of each actuator according to these these data.Among these three operating conditions, the first operating condition of particular importance and the second operating condition are engine speed and engine loading, and the 3rd operating condition is engine water temperature.If determined engine speed, engine loading and engine water temperature, can determine the control desired value that can make the operating condition of engine be placed in best each actuator.This determines to utilize engine is carried out to actual tests and the proper data that obtains.But, if whole combination of engine speed, engine loading and engine water temperature is obtained to suitable data, will in suitableization operation, expend huge time and man-hour.Therefore propose to be, only certain specific combination of engine speed, engine loading and engine water temperature is obtained to proper data, other combination is implemented to interpolation according to obtained proper data and calculate.The engine control system of present embodiment has implemented to utilize the computing of the actuator control desired value of interpolation calculating concurrently by polycaryon processor 100.
The proper data using in polycaryon processor 100 is prepared with the combination of the engine speed shown in lattice point, engine loading and engine water temperature in for Fig. 4.Fig. 4 is to represent the figure of the arrangement of the lattice point on the three-dimensional cartesian coordinate system taking engine speed as X-axis, taking engine loading as Y-axis, taking engine water temperature as Z axis.On this three-dimensional cartesian coordinate system, be arranged with 100 lattice points in the mode of 4 layers of 5 row 5 row.On three-dimensional cartesian coordinate system, the three-dimensional dot pattern being obtained by lattice point is simple cubic lattice, has defined an apron space by eight lattice points of place, the cubical summit configuration as unit dot matrix.At each lattice point place, associated respectively have with respect to engine speed, engine loading and the engine water temperature of described position and the proper data of suitably being changed.The proper data at each lattice point place is programmed in Optimal Control value operation program, and is assigned to and has been set up associated core 102 with this lattice point.
Fig. 5 is to represent to be arranged on the associated figure between the core 102 on lattice point and the polycaryon processor 100 on the three-dimensional cartesian coordinate system shown in Fig. 4.Being equipped on incidental numeral on a part of core among the core 102 on polycaryon processor 100 is, has been established the coordinate numbering of associated lattice point with described core.Accordingly, the lattice point that be accompanied with that (2,1,3) this digital core and X coordinate are 2, Y coordinate is 1, Z coordinate is 3 is established associated.In Fig. 5,100 lattice points shown in pictorial image 4 is inwhole, and only illustrate 4 layers of 3 row 3 row that approach initial point wherein amount to the associated of 36 lattice points and 36 cores.In Fig. 4, from each core 102, the coordinate of incidental lattice point numbering can be found out, lattice point is established associated with core 102 in 1 to 1 mode.In addition, the relation between the number of the core 102 having according to lattice point number and polycaryon processor 100, also exists residue to have the situation of not setting up associated core 102 with lattice point.But, in the following description, in the situation that mentioning specific core, there will be to use and set up the situation of the coordinate numbering of associated lattice point with described core.
In addition, be configured in the situation on three-dimensional cartesian coordinate system with respect to lattice point, because core 102 is configured on two-dimensional direct angle coordinate system, therefore the arrangement of lattice point merely cannot be replaced to the arrangement of nucleation 102.In present embodiment, according to fixing rule, and implement the association to the core 102 on two-dimensional direct angle coordinate system of lattice point on three-dimensional cartesian coordinate system.Fig. 6 is the figure for described rule is elaborated.Below, utilize Fig. 6, the lattice point on the three-dimensional cartesian coordinate system adopting in present embodiment is described to the associated method of the core 102 in plane.
In Fig. 6, the coordinate that is similarly accompanied with lattice point with Fig. 5 on each core 102 is numbered, and, be painted with the frame that surrounds four adjacent cores.Described frame represents, has formed the implication of a group by four cores 102 that surrounded by it.The incidental symbol of centre at frame is the identiflication number of described core group, and is set up the numbering of the content of having found associated lattice point for representing with described core.In each core group, be associated with the lattice point of different four connections of each coordinate coordinate identical and on Z axis in X-axis and Y-axis.For example, associated in the core group of identiflication number Gr.11 have, and X coordinate is 1, Y coordinate is 1, four lattice points of Z coordinate from 1 to 4.These four lattice points start in turn to be deasil assigned to each core taking the core of the lower left in core group as initial point from the lattice point of the numeral minimum of Z coordinate.And, each core group with taking engine speed as X-axis, arrangement mode that the lattice point of engine loading on the two-dimensional direct angle coordinate system of Y-axis is identical is arranged., the identiflication number of core group also becomes the coordinate of described core group itself.For example identiflication number Gr.32 represents, is the 3rd, in Y direction, is the core group of second in X-direction.
In being established associated core 102 with lattice point, be assigned with, used with respect to engine speed, engine loading and the engine water temperature of described lattice point and the Optimal Control value operation program of the proper data suitably changed.Optimal Control value operation program is installed in local storage 106 together with interpolation calculation procedure.Optimum value operation program is, uses engine characteristics model and the program of the Optimal Control value of actuator being carried out to computing.Described applicable data are used in the parameter value of engine characteristics model.In the polycaryon processor 100 of present embodiment, only have to be established associated core 102 with lattice point and to bring into play function as the optimum value operation core of the computing for actuator control desired value, core is in addition responsible for other processing required in engine control.
Next, utilize Fig. 7 and Fig. 8, the operational method that has used control desired value polycaryon processor 100, actuator is described.In the three-dimensional cartesian coordinate system shown in Fig. 7, some A represents, the operating point that determined by each currency of engine speed, engine loading and engine water temperature, i.e. current operating point.Said current referring to herein, it is current that the start time of this time step puts.For control engine rightly, need to carry out computing to the Optimal Control value of the actuator at current operating point place, and be set the control desired value into actuator.But in three-dimensional cartesian coordinate system, the lattice point of having prepared the proper data for Optimal Control value being carried out to computing is only aligned to the lattice point of 100 of 4 layers of 5 row 5 row, does not prepare proper data in region in addition.Therefore, cannot directly calculate by Optimal Control value operation program the Optimal Control value at current operating point place.
Therefore, in the present embodiment, the Optimal Control value at eight lattice point places that surround current operating point is carried out to computing, and utilize these Optimal Control values and the Optimal Control value at current operating point place is implemented to interpolation and calculate.In the example depicted in fig. 7, current operating point A belongs to by coordinate (2,2,2), coordinate (2,3,2), coordinate (3,2,2), coordinate (2,2,3), coordinate (2,3,3), coordinate (3,2,3), coordinate (3,3,2) and the lattice point of coordinate (3,3,3) and the space that is defined.Therefore, if the Optimal Control value at known these adjacent eight lattice point places can be calculated the Optimal Control value that obtains current operating point A place by the interpolation that has utilized these Optimal Control values.
The Optimal Control value at current operating point A place is specifically calculated by following order.First,, in the configuration of the core shown in Fig. 8, polycaryon processor 100 will be established (2,2 among associated core 102 with lattice point, 2), (2,3,2), (3,2,2), (2,2,3), (2,3,3), (3,2,3), (3,3,2) and the core of (3,3,3) be applied in computing.These eight cores are respectively by being installed in the Optimal Control value operation program in self local storage 106, and the Optimal Control value of having been set up associated lattice point place with self is carried out to computing.The computing of the Optimal Control value of now, being implemented by these eight cores is implemented concurrently.In the time having completed the computing of Optimal Control value by these eight cores, the data of its operation result will be concentrated to eight representative core places among core.Each core 102 is programmed as follows,, be identified as and represent lattice point surrounding lattice points among adjacent eight lattice points of current operating point, approach initial point most on three-dimensional cartesian coordinate system, and will represent that lattice point has been established associated core and has been identified as and represents core with described.As shown in Figure 8, the core of (2,2,2) becomes and represents core.As representing (2 of core, 2,2) core utilizes the data of the Optimal Control value self calculating and the data of the Optimal Control value that receives from other seven cores, by being installed in the interpolation calculation procedure in self local storage 106, and the Optimal Control value at current operating point A place is carried out to interpolation calculating.
According to above-mentioned method, the Optimal Control value at required adjacent eight the lattice point places core by separately and by computing concurrently respectively in order to carry out interpolation calculating.Thus, compared with utilizing the situation of existing single core processor, can shorten significantly the needed time in the computing of control desired value of actuator.
In addition, by implement associated between lattice point and core with above-mentioned rule, thereby can by physically also eight cores close to each other carry out the computing of the Optimal Control value at eight adjacent lattice point places.And one in these eight cores is calculated core and brings into play function as interpolation.Set according to these, not only can reduce the poor of in the time that the data of the Optimal Control value calculating in each core are calculated to core and transmitted to the interpolation communication delay that produces, and can make communication delay itself reduce.
In addition, above-described method is implemented by the Optimal Control value calculation procedure shown in the process flow diagram of execution graph 9 in each core 102.This program is by being installed in the application program in the local storage 106 of each core 102, and for example, is performed with regular time step (control cycle of engine) in each core 102.
In the initial step S102 of the process flow diagram of Fig. 9, core 102 is obtained each currency of engine speed, engine loading and engine water temperature.These values are calculated by the sensor-based output valve of core beyond Optimal Control value operation core.Then, core 102 is determined current operating point according to each currency of engine speed, engine loading and engine water temperature, and judges that whether current operating point is in having been set up associated lattice point with self in interior eight adjacent defined spaces of lattice point by comprising.
In the case of the result judging of step S102 be negative, core 102 stop Optimal Control value computing and finish the computing in this time step.This is in order to suppress meaningless power consumption.
In the case of the result judging of step S102 be sure, core 102 performs step the processing of S104.In step S104, core 102 is carried out Optimal Control value operation program, and the Optimal Control value of having been set up associated lattice point place with self is carried out to computing.
In following step S106, core 102 represents that to self whether belonging to core judges.Be whether to represent that nuclear energy is enough to judge according to the position relationship of having been set up with self between associated lattice point and current operating point.
In the case of the result judging of step S106 be negative, core 102 performs step the processing of S112.In step S112, the data of the Optimal Control value that core 102 calculates self are to representing that core sends.Which core belongs to represents that nuclear energy enough judges according to the coordinate of the current operating point on three-dimensional cartesian coordinate system.The transmission of the data by Optimal Control value, thereby the computing being through with in this time step of being implemented by this core 102.
In the case of the result judging of step S106 be sure, core 102 performs step the processing of S108.In step S108, core 102 from operating point to current under other the lattice point that defines of operating space be established seven cores of associated other, receive the data of the Optimal Control value being gone out by these kernel operations.
In following step S110, according to comprising the Optimal Control value that himself calculates in Optimal Control value interior, adjacent eight lattice point places, by interpolation calculation procedure, the Optimal Control value at current operating point place is carried out to interpolation calculating as the core 102 that represents core.The Optimal Control value of calculating the current operating point place obtaining by interpolation is output as the control desired value of actuator.Accordingly, the computing being through with in this time step of implementing by this core 102.
By carrying out above program by each core 102, thereby can in the control desired value that can calculate at high speed actuator, suppress power consumption.
In addition, the feature of the engine of using as the applicable automobile of control device of the present invention, can be enumerated as the different this point of operating frequency under every kind of operating condition.For example, be generally shorter than the duration of runs before warming-up between the running after warming-up, be shorter than the duration of runs of full load the duration of runs of idling.This represents, also has the space that space that operating frequency is higher and operating frequency are lower on the three-dimensional cartesian coordinate system taking engine speed, engine loading and engine water temperature as axle.In the case, the lattice point in the space higher with belonging to operating frequency has been established associated core and other nuclear phase ratio, and its computing load is relatively high.If computing load concentrates on specific core, can cause the deteriorated of this core, and can make the lost of life of polycaryon processor 100 entirety.
Therefore,, in the engine control system of present embodiment, as shown in figure 10, implemented the processing that the responsible lattice point of each core is replaced successively in core group., carried out the rotation of the distribution of core.Although in the example depicted in fig. 10, rotation is deasil implemented in the mode shown in arrow, also can implement widdershins.
Distribute rotation program by carry out the core shown in the process flow diagram of Figure 11 in each core 102, thereby implemented the rotation of the distribution of core.In the initial step S202 of this program, whether pass through the schedule time duration of runs of the engine to the accumulative total from the enforcement of the distribution rotation of last core to judge above.In the case of this result judging be sure, execution step S204 processing.In step S204, the moment that the ignition switch in the time that this running starts is opened is implemented the distribution rotation of core.By implement by this way the distribution rotation of core according to each predetermined operation process, thereby can prevent because computing load concentrates on the deteriorated of the core that causes on a part of core.And this program is performed by the application program being installed in the local storage 106 of each core 102.
Other
The present invention is not limited to above-mentioned embodiment, can implement without departing from the spirit and scope of the invention various changes.For example can change in such a way and implement.
In the above-described embodiment, be to implement interpolation by chosen representative core from the core that calculates Optimal Control value to calculate.But, also can in other cores, implement interpolation and calculate.For example, Figure 12 and Figure 13 illustrate respectively the Change Example of above-mentioned embodiment.As shown in the Change Example of Figure 12, any one core being established beyond associated core 102 can be made as to interpolation calculating core with lattice point.Or, as shown in the Change Example of Figure 13, also can make to be established with lattice point any one among associated core 102 and be fixed as interpolation and calculate core.In these Change Examples, in the time the data of the Optimal Control value calculating in each core being sent to interpolation calculating core, thereby can become large corresponding to its distance communication delay physically.But because the computing of the Optimal Control value using in interpolation is calculated is to implement in the core approaching physically, therefore the difference of internuclear communication delay is less.Because the difference of communication delay is less, thereby can carry out with high-level efficiency computing as polycaryon processor entirety.
Setting up associated core with lattice point can be both that a part that is equipped on the core in polycaryon processor can be also whole.If set up associatedly with lattice point on the whole core that is equipped on polycaryon processor, in engine control, the processing of required other only need be responsible for by being established the less core of average load in associated core with lattice point.In addition, the quantity that is equipped on the core on polycaryon processor is preferably more than four, and more preferably more than nine, this quantity does not have concrete restriction.In addition, be arranged on the quantity of the lattice point on three-dimensional cartesian coordinate system as long as be equipped on below the quantity of the core on polycaryon processor, this quantity does not have concrete restriction.
In the control desired value of decision actuator, among three operating conditions of institute's foundation, the 3rd operating condition can change according to the state of engine.For example can be in the following way, that is, till the warming-up of engine finish during in, identical with above-mentioned embodiment and use engine water temperature as the 3rd operating condition, and after the warming-up of engine finishes, intake air temperature is used as the 3rd operating condition.Even in the case of the warming-up of engine finishes, intake air temperature also can change.Therefore,, compared with engine water temperature, the impact under the situation that intake air temperature has finished at warming-up, the state of engine being caused is larger.Therefore, by after finishing at the warming-up of engine, the 3rd operating condition is changed to intake air temperature from engine water temperature, thereby can more effectively apply in a flexible way the three-dimensional arrangement of lattice point the and more accurately state of engine being controlled.The 3rd operating condition after the warming-up of engine finishes also can replace intake air temperature and be made as the speed of a motor vehicle.This is because the intensity of the wind that travels according to the difference of the speed of a motor vehicle can be different, thereby can produce difference in the temperature of exhaust system.In addition, if the operating condition changing continuously is preferably made as the 3rd operating condition.For example air pressure and EGR lead and are suitable for as the 3rd operating condition.
Symbol description
100 ... polycaryon processor;
102 ... core;
104…CPU；
106 ... local storage;
108 ... bus;
110…OS；
112 ... application program.
Claims (9)
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| PCT/JP2011/078691 WO2013088494A1 (en) | 2011-12-12 | 2011-12-12 | Engine control device |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN103999073A true CN103999073A (en) | 2014-08-20 |
| CN103999073B CN103999073B (en) | 2017-04-05 |
Family
ID=48611984
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201180075464.3A Expired - Fee Related CN103999073B (en) | 2011-12-12 | 2011-12-12 | Engine control system |
Country Status (5)
| Country | Link |
|---|---|
| US (1) | US9442472B2 (en) |
| EP (1) | EP2793141B1 (en) |
| JP (1) | JP5704254B2 (en) |
| CN (1) | CN103999073B (en) |
| WO (1) | WO2013088494A1 (en) |
Families Citing this family (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US9624774B2 (en) * | 2011-09-28 | 2017-04-18 | Toyota Jidosha Kabushiki Kaisha | Engine control apparatus |
| JP2015109071A (en) * | 2013-10-25 | 2015-06-11 | トヨタ自動車株式会社 | Control device |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US5649198A (en) * | 1993-02-19 | 1997-07-15 | Fujitsu Limited | Mapping calculation units by dividing a calculation model which can be calculated in parallel on an application program |
| JP2009215944A (en) * | 2008-03-10 | 2009-09-24 | Hitachi Ltd | Electronic control system and method for operating same |
| EP2341448A1 (en) * | 2009-12-29 | 2011-07-06 | Robert Bosch GmbH | An electronic control unit and a method of performing interpolation in the electronic control unit |
Family Cites Families (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JPS62240441A (en) | 1986-04-09 | 1987-10-21 | Hitachi Ltd | fuel control device |
| SE509805C2 (en) * | 1994-08-11 | 1999-03-08 | Mecel Ab | Method and system for the control of internal combustion engines |
| JP3653841B2 (en) | 1996-01-23 | 2005-06-02 | 株式会社日立製作所 | Problem area division / allocation method |
| US6304671B1 (en) * | 1997-10-08 | 2001-10-16 | Seiko Epson Corporation | Image processing apparatus using an offset correction table enabling interpolation to be performed without conditional jumps |
| JP3341688B2 (en) * | 1997-10-08 | 2002-11-05 | セイコーエプソン株式会社 | Control method for image processing apparatus and image processing apparatus controlled by the method |
| JPH11280519A (en) * | 1998-03-30 | 1999-10-12 | Suzuki Motor Corp | Fuel injection engine |
| US6519553B1 (en) * | 1998-05-14 | 2003-02-11 | Sandia Corporation | Multiprocessor computer overset grid method and apparatus |
| JP4845761B2 (en) | 2007-02-09 | 2011-12-28 | 富士通株式会社 | Three-dimensional model shape simplification device and three-dimensional model shape simplification program |
| JP4687685B2 (en) | 2007-04-24 | 2011-05-25 | 株式会社デンソー | Electronic control device for engine control and microcomputer |
| JP5540609B2 (en) | 2009-09-01 | 2014-07-02 | 富士通株式会社 | Parallel computing system and communication control program |
| JP5321691B2 (en) * | 2009-11-16 | 2013-10-23 | 富士通株式会社 | Parallel computing device, parallel computing method, and parallel computing program |
| US8738860B1 (en) * | 2010-10-25 | 2014-05-27 | Tilera Corporation | Computing in parallel processing environments |
| US8566002B2 (en) * | 2011-04-18 | 2013-10-22 | GM Global Technology Operations LLC | Engine control systems and methods |
-
2011
- 2011-12-12 CN CN201180075464.3A patent/CN103999073B/en not_active Expired - Fee Related
- 2011-12-12 US US14/364,388 patent/US9442472B2/en not_active Expired - Fee Related
- 2011-12-12 WO PCT/JP2011/078691 patent/WO2013088494A1/en active Application Filing
- 2011-12-12 EP EP11877581.6A patent/EP2793141B1/en not_active Not-in-force
- 2011-12-12 JP JP2013548971A patent/JP5704254B2/en not_active Expired - Fee Related
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US5649198A (en) * | 1993-02-19 | 1997-07-15 | Fujitsu Limited | Mapping calculation units by dividing a calculation model which can be calculated in parallel on an application program |
| JP2009215944A (en) * | 2008-03-10 | 2009-09-24 | Hitachi Ltd | Electronic control system and method for operating same |
| EP2341448A1 (en) * | 2009-12-29 | 2011-07-06 | Robert Bosch GmbH | An electronic control unit and a method of performing interpolation in the electronic control unit |
Also Published As
| Publication number | Publication date |
|---|---|
| CN103999073B (en) | 2017-04-05 |
| US9442472B2 (en) | 2016-09-13 |
| EP2793141A4 (en) | 2015-10-28 |
| WO2013088494A1 (en) | 2013-06-20 |
| EP2793141B1 (en) | 2018-02-28 |
| EP2793141A1 (en) | 2014-10-22 |
| US20140343823A1 (en) | 2014-11-20 |
| JP5704254B2 (en) | 2015-04-22 |
| JPWO2013088494A1 (en) | 2015-04-27 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Wang et al. | Demand side management of plug-in electric vehicles and coordinated unit commitment: A novel parallel competitive swarm optimization method | |
| US20130054110A1 (en) | Control device for internal combustion engine | |
| JP2013228859A (en) | Plant control device | |
| CN103999073A (en) | Engine control device | |
| CN103917965B (en) | Engine control system | |
| CN102831230B (en) | A kind of parts selection method and system of hybrid power assembly | |
| US11453307B2 (en) | Control method for vehicle, information processing device, and vehicle control system | |
| Liu et al. | Multi-robot task allocation based on utility and distributed computing and centralized determination | |
| CN110518644B (en) | Reactive power optimization method and device for wind power plant and wind power plant | |
| CN116581827A (en) | A start-stop strategy for wind farm units under load-limited conditions considering the influence of wake turbulence | |
| JP7482896B2 (en) | Internal Combustion Engine Controller | |
| CN107748493A (en) | TBCC dcs communication network mixed topology structural optimization methods | |
| JP5853752B2 (en) | Control device for internal combustion engine | |
| JP2020181244A (en) | Electronic control device | |
| Lin et al. | An Offloading Strategy for Ocean Edge Computing Based on Whale Optimization Algorithm | |
| US20230294718A1 (en) | Distributed embedded controller implementation for self-learning controls | |
| JP5742684B2 (en) | Engine predictive control device | |
| Ganesh Kumar et al. | Fuzzy logic and neural network based induction control in a diesel engine | |
| CN118862579A (en) | A battery pack structure optimization method | |
| CN114511156A (en) | A method and device for optimizing ordered charging with partial disordered charging | |
| IT202100020744A1 (en) | Method of modeling a powertrain and controlling the modeled powertrain | |
| CN118508436A (en) | Multi-objective optimal scheduling method and device for micro-grid | |
| Wani et al. | Hybrid Neural Network Based Model for Predicting the Performance of a Two Stroke Spark Ignition Engine | |
| JP2014051911A (en) | Engine control system | |
| JP2014066165A (en) | Engine control device |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| CF01 | Termination of patent right due to non-payment of annual fee | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20170405 Termination date: 20191212 |